# Research Papers — 2026-05 · bundle page 110 of 326 > Full-text markdown of papers #5451–5500 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-05/page/110 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/109.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/111.md) ### Compiling Agentic Workflows into LLM Weights: Near-Frontier Quality at Two Orders of Magnitude Less Cost Authors: Simon Dennis, Rivaan Patil, Kevin Shabahang, Hao Guo Published: 2026-05-21 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.22502v1 Agent orchestration frameworks have proliferated, collectively exceeding 290,000 GitHub stars across LangGraph, CrewAI, Google ADK, OpenAI Agents SDK, Semantic Kernel, Strands, and LlamaIndex. All follow the same pattern: an external orchestrator above the LLM, injecting instructions and routing decisions every turn. Recent work has shown this architecture is dominated for procedural tasks by simply providing the procedure in a frontier model's system prompt (Dennis et al., 2026a), at the cost of consuming the context window, requiring a frontier model for every conversation, and exposing proprietary procedures to third-party providers. Compiling the procedure into the weights of a small fine-tuned model -- creating a subterranean agent -- should resolve all of these concerns, and prior work (SimpleTOD, FireAct, SynTOD, WorkflowLLM, Agent Lumos) has shown the technique works. Yet developer adoption has overwhelmingly favored orchestration. We identify three perceived barriers and address each empirically across travel booking (14 nodes), Zoom support (14 nodes, product-specific knowledge), and insurance claims (55 nodes, 6 decision hubs). --- ### BeLink: Biomedical Entity Linking Meets Generative Re-Ranking Authors: Darya Shlyk, Stefano Montanelli, Lawrence Hunter Published: 2026-05-21 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2605.22501v1 Despite recent progress, Biomedical Entity Linking (BEL) with large language models (LLMs) remains computationally inefficient and challenging to deploy in practical settings. In this work, we demonstrate that instruction-tuning of open-source generative models can offer an effective solution when applied at the re-ranking stage of the BEL pipeline. We propose a set-wise instruction-tuning formulation that enables fast and accurate candidate selection. Our method demonstrates strong performance on multiple BEL benchmarks, yielding significant improvements in linking accuracy (3%-24%) while reducing inference time compared to the state-of-the-art. We integrate our generative re-ranker into BeLink, a modular, end-to-end system designed for practical real-world BEL applications. --- ### The Neural Compiler: Program-to-Network Translation for Hybrid Scientific Machine Learning Authors: Lucas Sheneman Published: 2026-05-21 Categories: cs.LG, cs.AI, cs.SC arXiv: https://arxiv.org/abs/2605.22498v1 Scientific machine learning often requires combining known physics with unknown parameters or correction terms learned from data. Existing approaches either ignore known structure, encode it as a soft penalty, or require hand-written PyTorch code for each equation. We present The Neural Compiler, a system that translates programs written in a first-order Scheme-like expression language into frozen, differentiable PyTorch modules. These modules match the source program to floating-point precision and provide gradients through autograd. In hybrid models, the compiled module encodes known physics exactly while learned components model the unknown remainder. We evaluate the compiler across six experiment domains: Feynman physics equations, Lotka-Volterra dynamics, a damped pendulum, a one-dimensional heat equation, three-dimensional vector mechanics, and compositional generalization. Compiled modules match hand-coded PyTorch implementations numerically for single equations, showing no accuracy loss from compilation. With only 1 to 4 trainable parameters, compiled models recover physical constants to less than 1 percent error in most cases, while standard PINN baselines with more than 8500 parameters show 7 to 93 percent error. Compiled modules also compose with zero error, while neural approximations can accumulate large errors in deep composition chains. The main value of the compiler is not improved accuracy over hand-coded equations, but systematic composability: it generates correct, differentiable modules from symbolic specifications without rewriting each equation by hand. The system supports 51 primitive operations, including vector and matrix algebra, enabling PDE discretizations and hybrid scientific models. This string-in, module-out interface also provides a natural target for large language models that translate scientific descriptions into executable differentiable modules. --- ### The Signal in the Noise: OOD Detection Through Goodness-of-Fit Testing in Factorised Latent Spaces Authors: Philipp Bomatter, Jack Geary, Henry Gouk Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22496v1 Deep generative models offer a natural foundation for out-of-distribution (OOD) detection, yet prior work has shown that their assigned likelihoods are notoriously unreliable indicators for in- vs out-of-distribution data. In this paper, we address this problem by leveraging the diffeomorphic and mass-preserving properties of continuous normalising flows. Our analysis shows that OOD samples are mapped to noise samples that are highly atypical under the noise prior in ways not captured by the likelihood. Based on this observation, we propose a new method -- Signal in the Noise (SITN) -- for OOD detection on the single-sample level. SITN requires no access to OOD data, incurs minimal computational overhead, and provides strict control of false positive rates. Comprehensive evaluations through standard benchmarks and synthetic perturbations highlight the method's effectiveness and the absence of the complexity bias inherent to likelihood-based methods. --- ### Understanding Multimodal Failure in Action-Chunking Behavioral Cloning Authors: Lorenzo Mazza, Massimiliano Datres, Ariel Rodriguez, Sebastian Bodenstedt, Gitta Kutyniok, Stefanie Speidel Published: 2026-05-21 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2605.22493v1 Behavioral cloning becomes difficult when the same observation admits several valid actions. We study this problem for action-chunking policies and show that different multimodal parameterizations fail in different ways. For latent-variable policies, posterior-prior regularization makes deployment-time sampling more reliable, but excessive regularization removes the action-conditioned information needed to distinguish demonstrated modes. Reducing this regularization can preserve mode information, but then success depends on whether the prior covers the relevant latent regions. For action-space generative policies, multimodality is constrained by the smoothness of the base-to-action transport: a map with small Lipschitz constant cannot assign substantial probability to many well-separated modes. Covering many modes therefore requires either sharp transitions in base space or off-support bridge regions in action space. Experiments on synthetic multimodal tasks and robotic simulation benchmarks support these mechanisms. --- ### Training-Free Fine-Grained Semantic Segmentations in Low Data Regimes: A FungiTastic Baseline Authors: Sebastian Cavada, Francesco Pelosin, Lapo Faggi Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22492v1 Fine-grained semantic segmentation requires both precise localization and discrimination between visually similar classes. In FungiTastic, this problem is further complicated by a long-tailed distribution and strong variation in image acquisition conditions. We propose a training-free two-stage framework that decouples segmentation from classification. SAM3 first produces class-agnostic mushroom masks using macro-taxonomic prompts, and DINOv3 then assigns fine-grained labels through prototype matching in the embedding space. To improve this stage, we apply a simple transformation of the DINOv3 feature space that improves prototype-based classification. Compared with class-specific prompting, our approach is more scalable and keeps the segmentation cost low. We report results from one-shot to few-hundred-shot regimes, providing, to the best of our knowledge, the first baseline for fine-grained semantic segmentation in low-data settings. --- ### Represented Is Not Computed: A Causal Test of Candidate Algorithmic Intermediates in a Transformer Authors: Ishita Darade, Sushrut Thorat Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22488v1 Structured prompts require integrating components according to task-relevant relations. How a network implements this integration is often hard to judge in language or vision, where those relations are rarely specified precisely enough to define a candidate internal algorithm. Arithmetic offers a cleaner setting. We study a Transformer trained on base-digit extraction: given $N$, $B$, and $D$, it must report the coefficient of $B^D$ in the base-$B$ expansion of $N$. The closed-form solution, $\lfloor N/B^D \rfloor \bmod B$, provides explicit candidate algorithmic intermediates. Across three seeds, the model reaches 99.83% exact-answer accuracy on held-out number-base intersections, establishing reliable task competence. Linear probes decode the intermediates, making staged arithmetic computation plausible. Causal tests then separate representation from use: within the localized route from the stream with $D$ as input to the output positions, behavior depends on early $D$-selective communication, independent of $N$ and $B$. Relatedly, a sparse circuit search finds mostly separate $N$, $B$, and $D$ routes that combine late rather than the staged route suggested by the probes. Thus, the model represents the intermediates that make the closed-form solution plausible, but the identified localized causal route does not transmit them to the output stream. This case shows that probe-based conclusions can diverge sharply from causal observations, even when explicit algorithmic hypotheses are available. --- ### Polite on the Surface, Broken in Practice: A Curated Dataset for Fixing Generation and Register Failures in Low-Resource Bangla Text Generation Authors: Md. Asaduzzaman Shuvo, Mahedi Hasan, Md. Tashin Parvez, Azizul Haque Noman, Md. Shafayet Hossain Ovi Published: 2026-05-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.22487v2 Recent advances in Multilingual Large Language Models (MLLMs) have significantly enhanced cross-lingual conversational capabilities, yet modeling culturally nuanced and context-dependent communication remains a critical bottleneck. Specifically, existing state-of-the-art models exhibit a severe pragmatic gap when handling structural variations, regional idioms, and honorific consistencies in low-resource contexts like Bangla. To address this limitation, we introduce a novel, culturally aligned instruction-tuning dataset for \textbf{BangLa Application and DialoguE generation - BLADE} and benchmarking framework comprising $4,196$ meticulously curated interaction pairs. We leverage this resource to systematically fine-tune and evaluate leading open-weight architectures, including DeepSeek-8B and LLaMA-3.2-3B, utilizing parameter-efficient fine-tuning via LoRA adapters in a 4-bit NormalFloat (NF4) quantization framework. Our empirical evaluations demonstrate that models fine-tuned on our dataset yield substantial improvements in structural fidelity and honorific alignment, providing a rigorous benchmark for bridging pragmatic disparities in low-resource multilingual text generation. Code and dataset: https://github.com/ashuvo25/Bangla_Application_LLM/tree/main --- ### Supervised Classification Heads as Semantic Prototypes: Unlocking Vision-Language Alignment via Weight Recycling Authors: David Méndez, Roberto Confalonieri, Natalia Díaz Rodríguez Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22484v1 Vision-Language Models (VLMs) excel at tasks like zero-shot classification and cross-modal retrieval by mapping images and text to a shared space, but this requires expensive end-to-end training with massive paired datasets. Current post-hoc alignment methods reduce computational costs by connecting pretrained encoders through lightweight mappings, yet still demand substantial paired data. In this work, we investigate the potential of repurposing the classification heads of pretrained vision models as semantic prototypes. The recycling of these weights, typically discarded after pretraining, unlocks two distinct capabilities: it enables zero-shot alignment by using weights as semantic anchors, and serves as a robust data augmentation strategy by mixing these prototypes with real image-text pairs. We demonstrate that integrating our approach with several state-of-the-art post-hoc alignment techniques consistently boosts accuracy in cross-modal retrieval, zero- and few-shot classification tasks. --- ### When Stronger Triggers Backfire: A High-Dimensional Theory of Backdoor Attacks Authors: Donald Flynn, Hadas Yaron Goldhirsh, Jonathan P. Keating, Inbar Seroussi Published: 2026-05-21 Categories: cs.LG, math.ST arXiv: https://arxiv.org/abs/2605.22481v1 Backdoor poisoning attacks behave counter-intuitively in high dimensions: stronger training triggers can help the defender. We study regularised generalised linear models on Gaussian-mixture data in the proportional regime ($p/n \to κ$), varying the training trigger strength $α$ against a fixed test trigger. Three phenomena emerge: (i) clean test accuracy increases with $α$; (ii) attack success peaks at a finite $α$ and then declines; and (iii) the most damaging trigger direction is the minimum eigenvector of the data covariance. We prove all three results in closed form for the squared loss, and extend (i) and (ii) to general convex GLM losses via a Gaussian-proxy fixed-point system. We identify a finite-sample noise floor proportional to $κ$ as the mechanism behind (i), invisible to classical $n \gg p$ analysis. Experiments on CIFAR-10 and Gaussian surrogates match the theory closely; ResNet-18 experiments show the same phenomena beyond the convex setting. --- ### Implicit Regularization of Mini-Batch Training in Graph Neural Networks Authors: Clement Wang, Antoine Vialle, Robin Vaysse, Thomas Bonald Published: 2026-05-21 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.22480v1 Mini-batch training of Graph Neural Networks (GNNs) is fundamentally different from training on i.i.d. data: sampling a subgraph alters the topology and introduces boundary effects, leading prior work to develop structure-aware samplers that preserve local connectivity and reduce embedding variance. Surprisingly, we demonstrate that the simplest possible scheme, Random Node Sampling (RNS), training on the induced subgraph of uniformly sampled nodes, matches or outperforms full-graph training on 8 of 10 datasets at a fraction of the wall-clock time and memory. To explain this, we apply backward error analysis to graph mini-batch Stochastic Gradient Descent (SGD) and show that it implicitly minimizes the sampled loss plus a regularizer proportional to the mini-batch gradient variance, a quantity directly shaped by the sampler. Although RNS discards local structure, it produces mini-batches whose expected loss is closer to the full-graph loss, and whose per-batch gradients have lower variance, yielding a better implicit objective. Our analysis reframes the choice of graph sampler as a form of implicit regularization, and identifies RNS as a strong, theoretically grounded method for scalable GNN training. --- ### DeliCIR: Memory-Guided Test-Time Deliberation via Multi-Agent Collaboration for Composed Image Retrieval Authors: Xingtian Pei, Yukun Song, Changwei Wang, Shunpeng Chen, Rongtao Xu, Shengpeng Xu, Shibiao Xu Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22478v4 Composed Image Retrieval (CIR) requires both preserving the visual continuity of the reference image and faithfully executing the semantic variables specified in the modification text, which constitute the core challenge of the task. Existing methods often suffer from Perception Myopia in a single space, or fall into Logic Drift in iterative collaboration due to the perception ceiling of the underlying retriever. To address this issue, we propose a one-stop hierarchical Perception-to-Deliberation Framework (PDF), which, to the best of our knowledge, is the first to introduce Test-Time Scaling (TTS) into CIR. Relying on a hierarchical multi-agent architecture, PDF first utilizes an Intent Routing Manager to dynamically dispatch multi-view Worker perception signals based on modification intents to construct a high-recall candidate pool. Subsequently, the Deliberation Manager combines a training-free reasoning experience distillation mechanism with a tournament-style test-time scaling strategy (T-TTS) to perform experience-guided fine-grained reasoning and produce the final retrieval results. Experimental results demonstrate that PDF achieves SOTA performance on three benchmark datasets: CIRR, CIRCO, and Fashion IQ. This study indicates that experience-driven memory mechanism and TTS represent a highly promising and scalable path for achieving fine-grained multimedia retrieval. The code will be made publicly available upon acceptance. --- ### Structured-Sparse Attention for Entity Tracking with Subquadratic Sequence Complexity Authors: Hangyue Zhao, Paul Caillon, Erwan Fagnou, Alexandre Allauzen Published: 2026-05-21 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2605.22476v1 Entity tracking requires maintaining and updating latent states for entities and attributes over long sequences. Recent task-specific attention operators can compress deep Transformer stacks into a few layers by performing multi-hop state propagation within a single layer, but their dense evaluation remains expensive. We show that in this setting, learned attention is strongly structured: most mass concentrates in local block-diagonal neighborhoods with a light cross-block residue. Exploiting this, we derive a blockwise evaluation of a resolvent-style operator that keeps within-block interactions exact and routes cross-block interactions through a reduced system. The resulting evaluation is subquadratic in sequence length $O(n^{4/3}d)$ (and $O(n^{7/3})$ when $d\approx n$). On controlled tracking benchmarks, our method matches the dense operator's accuracy while reducing wall-clock time by $12-29\%$ under a standardized measurement protocol, and is up to $2.4 \times$ faster than a compact dense Transformer at comparable exact-match accuracy. We further provide ablations over block size and model capacity, and identify a limitation: performance collapses when the number of simultaneously evolving properties exceeds the number of attention heads. --- ### Winner-Take-All bottlenecks enforce disentangled symbolic representations in multi-task learning Authors: Julian Gutheil, Simon Hitzginger, Robert Legenstein Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22472v2 Winner-take-all (WTA) networks constitute a central circuit motif in cortical networks of the brain. In addition, WTA-like activations are abundant in modern deep learning models in the form of the softmax activation for example in attention layers of transformers. While their role in the extraction of latent factors has been studied for relatively simple generative models, their role in the context of highly non-linearly entangled latent factors has remained elusive. In this article, we show that a WTA bottleneck within a deep neural network can enforce under certain well-defined conditions the extraction of categorical latent factors of the data in a multi-task learning setup. In particular, we prove that the representation that emerges in the WTA bottleneck is highly symbolic, where a single neuron or a population of neurons encodes the presence of a single abstract feature such as a specific object, color, or position. We furthermore show empirically on two datasets, that this also holds for architectures and setups that do not fully comply with the assumptions of our theorem and demonstrate the advantages of the acquired symbolic representation for generalization. Our proposed model provides insights into the generalization capabilities of deep neural networks with WTA-like components and may serve as an interface between symbolic and subsymbolic AI systems. --- ### Lost in Tokenization: Fundamental Trade-offs in Graph Tokenization for Transformers Authors: Maya Bechler-Speicher, Gilad Yehudai, Gil Harari, Clayton Sanford, Amir Globerson, Joan Bruna Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22471v2 Transformers have become a central architecture for graph learning, but their application to graphs requires first choosing a tokenization: a graph-to-token map that determines which structural information is exposed at the input. In this work, we show that this choice is a fundamental component of transformer expressivity. We examine three tokenizations that serve as building blocks for many existing graph tokenizations: spectral, random-walk, and adjacency tokenizations. We prove that different tokenizations induce distinct depth regimes: the same graph computation may be realizable by a shallow transformer under one tokenization, while requiring substantially larger depth under another. For example, we prove that random-walk tokenization is lossy for any walk length, making it impossible in general to recover the graph from it, and that while spectral tokenization is lossless, it is ill-conditioned for local tasks. We further show that although both random-walk and spectral tokenizations are derived from adjacency information, it is impossible for a limited-depth transformer to convert between tokenization families in general. In particular, we establish lower bounds and impossibility results showing that unfavorable tokenizations may preclude the efficient recovery of more suitable structural representations. Finally, we complement our theory with controlled experiments on synthetic and real-world tasks, validating the predicted separations and showing that different tasks favor different structural views, and combining complementary tokenizations allows the transformer to leverage distinct signals from each representation. --- ### Lost in Tokenization: Fundamental Trade-offs in Graph Tokenization for Transformers Authors: Maya Bechler-Speicher, Gilad Yehudai, Gil Harari, Clayton Sanford, Amir Globerson, Joan Bruna Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22471v1 Transformers have become a central architecture for graph learning, but their application to graphs requires first choosing a tokenization: a graph-to-token map that determines which structural information is exposed at the input. In this work, we show that this choice is a fundamental component of transformer expressivity. We examine three tokenizations that serve as building blocks for many existing graph tokenizations: spectral, random-walk, and adjacency tokenizations. We prove that different tokenizations induce distinct depth regimes: the same graph computation may be realizable by a shallow transformer under one tokenization, while requiring substantially larger depth under another. For example, we prove that random-walk tokenization is lossy for any walk length, making it impossible in general to recover the graph from it, and that while spectral tokenization is lossless, it is ill-conditioned for local tasks. We further show that although both random-walk and spectral tokenizations are derived from adjacency information, it is impossible for a limited-depth transformer to convert between tokenization families in general. In particular, we establish lower bounds and impossibility results showing that unfavorable tokenizations may preclude the efficient recovery of more suitable structural representations. Finally, we complement our theory with controlled experiments on synthetic and real-world tasks, validating the predicted separations and showing that different tasks favor different structural views, and combining complementary tokenizations allows the transformer to leverage distinct signals from each representation. --- ### MaSC: A Masked Similarity Metric for Evaluating Concept-Driven Generation Authors: Patryk Bartkowiak, Lennart Petersen, Bartosz Kotrys, Dominik Michels, Soren Pirk, Wojtek Palubicki Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22469v1 Evaluating single-concept personalization in text-to-image diffusion requires measuring both concept preservation, which captures identity fidelity to a reference, and prompt following, which captures whether the generated scene matches the prompt. Existing metrics commonly compute these signals using global image or text-image embeddings, such as CLIP-I, DINO, and CLIP-T. We show that such metrics correlate poorly with human perception because they attend to the image as a whole instead of separating the concept subject from the background. We introduce MaSC, a masked similarity metric that uses externally provided foreground concept masks to decompose evaluation into subject-specific concept preservation and background-based prompt following. MaSC computes both scores from frozen SigLIP2 SO400M-NaFlex features: concept preservation is measured by masked max-cosine matching between foreground reference patches and generated-image patches, while prompt following is measured by comparing a background-only pooled image embedding to a subject-stripped prompt embedding. On DreamBench++ human ratings, MaSC achieves Krippendorff alpha = 0.471 for concept preservation, outperforming all tested non-LLM baselines and GPT-4V, and approaching GPT-4o. On ORIDa, a real-photo identity-preservation benchmark across physical environments, MaSC achieves AUC = 0.992, nearly perfectly distinguishing same-subject from cross-subject pairs. Its prompt-following score also outperforms the CLIP-T baseline shipped with DreamBench++. These results show that spatially decomposed aggregation is a strong design principle for evaluating concept-driven generation. --- ### BioFormer: Rethinking Cross-Subject Generalization via Spectral Structural Alignment in Biomedical Time-Series Authors: Guikang Du, Haoran Li, Xinyu Liu, Zhibo Zhang, Xiaoli Gong, Jin Zhang Published: 2026-05-21 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.22468v2 Cross-subject generalization in biomedical time-series refers to training on data from some subjects and testing on unseen subjects.The key challenge is to suppress subject specific variability in BTS representations.Most existing methods implicitly suppress the variability through model building or subject adversarial learning, but rarely model it explicitly.We introduce spectral drift as a new perspective to characterize subject specific variability.Specifically, BTS signals under the same label often share consistent oscillatory structure, yet exhibit subject-dependent magnitude or phase shifts in specific frequency components, which we interpret as subject-specific variability. Building on this insight, we propose BioFormer.At its core is a Frequency-Band Alignment Module(FBAM) that generates band-wise modulation factors from the spectral distribution and adaptively adjusts amplitude and phase to align spectral structure, thereby mitigating variability.We further pair FBAM with Sample Conditional Layer Normalization, which infers normalization parameters from intrinsic signal statistics rather than subject identity, stabilizing cross-subject representations.Extensive experiments on six datasets demonstrate that BioFormer outperforms 12 baselines, yielding absolute F1-score improvements of 6%. --- ### SADGE: Structure and Appearance Domain Gap Estimation of Synthetic and Real Data Authors: Patryk Bartkowiak, Bartosz Kotrys, Dominik Michels, Soren Pirk, Wojtek Palubicki Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22467v1 We propose SADGE, a quantitative similarity metric that predicts the performance of synthetic image datasets for common computer vision tasks without downstream model training. Estimating whether a synthetic dataset will lead to a model that performs well on real-world data remains a bottleneck in model development. Existing evaluation metrics (e.g., PSNR, FID, CLIP) primarily measure semantic alignment between real and synthetic images (Appearance Similarity Score). Less commonly, structural similarity between images is considered to assess the domain gap (Geometric Similarity Score). However, to the best of our knowledge there exists no studies that evaluate which similarity metric is the best downstream predictor for a given synthetic dataset. In this paper, we show over a wide variety of different synthetic datasets and downstream tasks that neither appearance nor geometry alone can reliably predict downstream performance; rather, it is their non-linear interplay that dictates synthetic data utility. Specifically, we measure how commonly used Appearance and Geometric Similarity metrics computed between synthetic and real images correlate with downstream performance in object detection, semantic segmentation, and pose estimation. Across five public synthetic-to-real benchmark families and 15 dataset-level variants (79k image pairs), SADGE achieves the strongest association with downstream transfer performance under both linear and rank-based criteria, reaching Pearson r=0.88 and Spearman rho=0.77. We compute for each combination of geometry-based methods and appearance-based approaches SADGE scores across all benchmark families. The best configuration is obtained by fusing DINOv3 appearance similarity with MASt3R geometric consistency through a constrained bilinear interaction, outperforming both the strongest geometry-only baseline and the strongest appearance-only baseline . --- ### In Silico Modeling of the RAMPHO Buffer: Dissociating Informational and Energetic Masking via Phonetic Entropy in Deep Neural Networks Authors: Stefan Bleeck Published: 2026-05-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.22465v1 The fundamental challenge of listening in multi-talker environments is a cognitive bottleneck, defined by the Ease of Language Understanding (ELU) model as a failure within the RAMPHO episodic buffer. Current deep neural networks for speech enhancement optimize purely for physical acoustics, failing to account for the cognitive penalty of informational masking. Here, we present an in silico simulation of the RAMPHO buffer using the frame-by-frame phonetic entropy of a self-supervised acoustic model (wav2vec 2.0). By contrasting a semantically intact distractor with a phase-decorrelated distractor (the Concentration Shield) across a signal-to-noise ratio (SNR) sweep, we successfully dissociate the cognitive penalty of informational distraction from the physical penalty of energetic decay. The simulation reveals a cognitive-acoustic Pareto optimization problem: destroying a distractor's semantic payload provides a release from informational masking at high SNRs, but fundamentally degrades temporal glimpsing cues at low SNRs. --- ### Reinforcement learning for ion shuttling on trapped-ion quantum computers Authors: Maximilian Schier, Lea Richtmann, Christian Staufenbiel, Tobias Schmale, Daniel Borcherding, Michèle Heurs, Bodo Rosenhahn Published: 2026-05-21 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2605.22463v1 Scalable trapped-ion quantum computing is commonly realized with modular chips that feature distinct zones with specific functionalities, such as storage, state preparation, and gate execution. To execute a quantum circuit, the ions must be transported between these zones. This process is called ion shuttling. To achieve reliable computation results, the shuttling process must be optimized. However, as the number of ions increases, this becomes a high-dimensional optimization problem where optimal solutions cannot be computed efficiently. We demonstrate, to the best of our knowledge, the first use of reinforcement learning (RL) for the optimization of ion shuttling. RL is well-suited for such scenarios, as it enables learning a strategy through direct interaction with the problem. We show that our RL approach outperforms current state-of-the-art heuristic techniques, yielding a reduction in shuttling operations of up to 36.3 %. Furthermore, we show that our method is easily applicable to various chip architectures. Our approach offers a versatile method to study shuttling efficiency during chip design and, therefore, a highly relevant tool for future, more complex architectures. --- ### From Correlation to Cause: A Five-Stage Methodology for Feature Analysis in Transformer Language Models Authors: Caleb Munigety Published: 2026-05-21 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.22462v1 We propose a five-stage methodology for causal feature analysis in transformer language models (probe design, feature extraction, causal validation, robustness testing, and deployment integration) and demonstrate it end-to-end on GPT-2 small performing the Indirect Object Identification (IOI) task. Activation patching recovers the canonical IOI circuit (layer-9 head 9 alone gives recovery +1.02). A sparse autoencoder recovers per-name selective features with effect sizes of 30 to 50 activation units. Causal validation finds these features specifically but only partially causal: ablating fifteen of them leaves the model accurate on 98% of prompts. Two NLA-inspired evaluations strengthen this picture: the fifteen selective features explain only 31% of activation variance versus the SAE's 99.7%, and selectivity ratio anticorrelates with causal force (r = -0.56). Robustness testing under three distribution shifts finds that the circuit transfers cleanly but feature ablation effects degrade substantially, exposing a gap between detection robustness and causal robustness. A cost-based deployment evaluation (assumed $50/FN, $0.42/FP, 2% error rate) finds an optimal monitor configuration yielding $8.96 per 1000 queries against a $1000 baseline, a 99.1% saving. Optimal composition strategy varies with cost ratio and base rate. The conjunction of stages produces findings no single stage would. --- ### KAPPS: A knowledge-based CPPS Architecture for the Circular Factory Authors: Etienne Hoffmann, Jan-Felix Klein, Sören Weindel, Max Goebels, Sebastian Behrendt, Daniel Hernández, Ratan Bahadur Thapa, Jürgen Fleischer, Kai Furmans, Steffen Staab Published: 2026-05-21 Categories: cs.AI, eess.SY arXiv: https://arxiv.org/abs/2605.22457v1 While linear manufacturing relies on homogeneous materials and predefined process sequences, circular manufacturing reintroduces used products with heterogeneous and uncertain conditions. This shift demands manufacturing systems capable of handling variable product states, dynamically reconfigurable processes, and the integration of human and machine knowledge. Conventional manufacturing IT architectures, designed for stable structures and deterministic execution, are unable to meet these requirements, as they cannot adequately represent and manage the uniqueness of individual components at runtime. Following a design science methodology for developing a Cyber Physical Production System for circular manufacturing, we derive 14 requirements from five complementary perspectives. Based on these requirements, we design KAPPS, a knowledge-based architecture that uses an ontology-grounded knowledge graph as a unifying data backbone, combined with a semantic interface layer to enable consistent data and information integration, reasoning, and communication across heterogeneous systems and services, turning the knowledge graph from an integration layer into the factories authoritative write-time state. KAPPS incorporates modules for constraint enforcement and event-driven planning, enabling incremental adaptation of execution plans under uncertainty and human-machine knowledge exchange. The applicability of KAPPS is demonstrated through two implemented use cases: (i) Anomaly detection and learning through knowledge graph mediated services and (ii) runtime constraint enforcement in a modular conveyor system. Subsequently, the architecture is evaluated against the 14 requirements (ed. abstract shortened) --- ### Steins;Gate Drive: Semantic Safety Arbitration over Structured Futures for Latency-Decoupled LLM Planning Authors: Anjie Qiu, Hans D. Schotten Published: 2026-05-21 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2605.22456v1 Cloud-hosted LLM driver agents provide useful semantic judgments, but their inference latency exceeds stepwise vehicle-control windows. Learned world models predict futures, but they usually keep future generation and action selection inside large coupled loops. We present SteinsGateDrive, a latency-decoupled planner-runtime architecture in which the worldline metaphor from the eponymous story names one plausible consequence of an intervention: the LLM selects counterfactual driving futures before the final control instant, and a runtime reuses the selected forecast only while safety contracts remain valid. The generator builds three world-line roles: alpha nominal ego-conditioned futures, beta interaction counterfactuals around nearby vehicles, and gamma hazard-stress futures such as braking, cut-ins, or blocked corridors. The selected branch becomes a typed StrategicForecast with horizon, validity/abort conditions, fallback, and authority. On a within-subject, matched-seed normal-highway protocol with 10 seeds and 20 steps, GPT-5.4 mini reduces effective lag from +3.07 s at 1-second horizon to -0.01 s at 4-second horizon while preserving the measured no-collision safety boundary. The architecture's safety contribution comes from the atom-predicate runtime check, not from the drift score, which functions as a refresh-frequency knob. --- ### Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light Authors: Valeria Pais, Malena Mendilaharzu, Daniele Faccio, Luis Oala, Christoph Clausen, Bruno Sanguinetti Published: 2026-05-21 Categories: cs.CV, cs.AI, cs.LG, physics.optics arXiv: https://arxiv.org/abs/2605.22455v2 Real-world deployment of AI vision models is both fueled and limited by the data available for training and testing. Real datasets are sparse and uneven: long-tailed or unbalanced distributions hinder generalization, and the low number of samples in low density regions makes it hard to run evaluations. Synthetic data can fill these gaps, providing us with a way to sample the input space more continuously and improve data coverage for benchmarks. Focusing on the autonomous driving safety-critical case of pedestrian detection in the dark, we show how synthetic low-light samples can be used to better characterize the performance of a state-of-the-art object detection model as a function of the scene illumination. We use a synthetic RAW image augmentation technique to generate low-light samples that match the noise model of the camera sensor. Performance metrics on real and synthetic low-light data are similar, indicating that the AI model finds it hard to distinguish between them. --- ### Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light Authors: Valeria Pais, Malena Mendilaharzu, Daniele Faccio, Luis Oala, Christoph Clausen, Bruno Sanguinetti Published: 2026-05-21 Categories: cs.CV, cs.AI, cs.LG, physics.optics arXiv: https://arxiv.org/abs/2605.22455v1 Real-world deployment of AI vision models is both fueled and limited by the data available for training and testing. Real datasets are sparse and uneven: long-tailed or unbalanced distributions hinder generalization, and the low number of samples in low density regions makes it hard to run evaluations. Synthetic data can fill these gaps, providing us with a way to sample the input space more continuously and improve data coverage for benchmarks. Focusing on the autonomous driving safety-critical case of pedestrian detection in the dark, we show how synthetic low-light samples can be used to better characterize the performance of a state-of-the-art object detection model as a function of the scene illumination. We use a synthetic RAW image augmentation technique to generate low-light samples that match the noise model of the camera sensor. Performance metrics on real and synthetic low-light data are similar, indicating that the AI model finds it hard to distinguish between them. --- ### Don't Forget the Critic: Value-Based Data Rehearsal for Multi-Cyclic Continual Reinforcement Learning Authors: Benjamin Poole, Andrew Quinn, Li Yang, Minwoo Lee Published: 2026-05-21 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.22454v1 Data rehearsal has emerged as a leading approach for mitigating catastrophic forgetting in Continual Reinforcement Learning (CRL). However, existing work remains confined to policy gradient frameworks, regularizing only actors due to the performance degradation incurred by critic regularization. This actor-centric approach overlooks the potential of data rehearsal for value function approximation. Moreover, existing evaluations in CRL rarely consider multi-cyclic environments where task sequences repeat, a critical real-world scenario that exacerbates forgetting and plasticity. We investigate data rehearsal for Deep Q-Networks using Q-value regularization in multi-cyclic settings and propose Qreg+NWLU which introduces two simple modifications: (1) continuous data rehearsal that dynamically collects and updates stored Q-values throughout training, and (2) "No-Wait" regularization that applies immediately rather than after the first task. Together, these modifications yield improvements in learning efficiency, forgetting mitigation, and knowledge transfer over Qreg and conventional CRL methods within value function approximation settings. --- ### S2ED: From Story to Executable Descriptions for Consistency-Aware Story Illustration Authors: Sijing Yin, Jiamou Liu, Xiao Tang, Yaser Shakib, Qian Liu Published: 2026-05-21 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.22448v1 Multi-frame story illustration requires long-horizon coherence beyond single-image text-to-image generation, including narrative decomposition and persistent character identity, layout, and affect across frames. We propose Story-to-Executable Descriptions (S2ED), a training-free, model-agnostic, prompt-layer framework that converts a full story into a sequence of explicit, editable executable descriptions for more consistent rendering. S2ED coordinates three agents to segment the narrative, ground canonical character attributes, and enrich spatial and affective cues, enabling interpretable prompt-carried state propagation and local edits to repair drift without retraining the generator. Experiments on Flintstones and Shakoo Maku show that S2ED improves sequence-level consistency and character fidelity over strong prompting, large-model planning, and a reference training-based method, under both automatic metrics and human judgments. We also deploy S2ED in an end-to-end story-to-storybook system for children's illustrated stories, with a supplementary video. --- ### Cohesion-6K: An Arabic Dataset for Analyzing Social Cohesion and Conflict in Online Discourse Authors: Aisha Ali Al-Athba, Wajdi Zaghouani Published: 2026-05-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.22447v1 The study of online discourse has become central to understanding societal polarization. While much research has focused on detecting overt toxicity, the subtle dynamics of social cohesion, meaning the interaction between divisive and unifying narratives, remain computationally underexplored (Bail, 2021; Gonzalez-Bailon and Lelkes, 2023). This paper presents Cohesion-6K, a manually and ChatGPT-assisted annotated dataset of six thousand Arabic public Facebook posts related to the Israeli Occupation of Palestine. Each post is assigned to one of five discourse categories that represent a continuum from conflict to cohesion: Conflict, Resolution, Community Engagement, Supportive Interactions, and Shared Values. The annotation process combines expert human judgment with model-assisted pre-labeling verified by trained annotators, achieving substantial inter-annotator agreement (Cohens kappa = 0.85). Quantitative analysis reveals a consistent engagement gap, where conflict-oriented posts receive between two and four times more user interaction than resolution-oriented ones (p < 0.01). This pattern illustrates how divisive discourse tends to attract disproportionate visibility in Arabic social media spaces. Cohesion-6K provides a transparent and reproducible resource for the study of online cohesion and polarization. The dataset, annotation guidelines, and preprocessing code will be released for research use under an open license, supporting future work in computational social science, digital communication, and Arabic natural language processing. --- ### Pre-VLA: Preemptive Runtime Verification for Reliable Vision-Language-Action and World-Model Rollouts Authors: Zhen Sun, Yongjian Guo, Haoran Sun, Luqiao Wang, Wei Lu, Jiachi Ji, Shengzhe Ji, Junwu Xiong, Zhijun Meng Published: 2026-05-21 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2605.22446v1 While large vision-language-action (VLA) models and generative world models (WM) have advanced long-horizon embodied intelligence, their practical deployment remains challenged by uncertainty in learning-based action generation. Low-quality actions may cause physical failures during execution or lead to misleading world-model rollouts with redundant rendering costs. To address this issue, we propose Pre-VLA, a unified runtime verification architecture that performs preemptive action validity assessment before physical execution or world-model imagination. Pre-VLA leverages an efficient multimodal backbone with modality-aware pooling and a lightweight dual-branch head to predict both safety confidence and critic-derived advantage scores for candidate action chunks. To handle severe class imbalance and unstable boundary decisions, we train Pre-VLA with a multi-task objective combining Focal classification, advantage regression, and soft-threshold calibration. During deployment, a dual-mode preemptive resampling scheduler filters low-quality actions and triggers adaptive resampling under a limited computation budget. Experiments on the LIBERO benchmark show that Pre-VLA improves the average closed-loop success rate across four suites from 30.79\% to 37.62\% over RynnVLA-002, reduces task execution steps, achieves 183.9 ms average forward verification time per action chunk, and mitigates error accumulation in world-model rollouts. --- ### A Constant-Time Implementation Methodology for Activation Functions on Microcontrollers Authors: Andrii Tyvodar, Andreas Rechberger, Dirmanto Jap, Shivam Bhasin, Bernhard Jungk, Jakub Breier, Xiaolu Hou Published: 2026-05-21 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2605.22441v1 Embedded neural-network inference can leak information through timing side channels, including leakage caused by the evaluation of activation functions. This work proposes a constant-time implementation methodology for activation functions on embedded microcontrollers and validates it on ReLU, sigmoid, tanh, GELU, and Swish on an ARM Cortex-M4 platform. The proposed methodology combines branchless selection, fixed-cost Padé-based approximation, dummy arithmetic where needed, and cycle alignment to obtain timing-regular activation-function implementations. As motivation, we also evaluate a desynchronization-based countermeasure and show that it remains vulnerable to a template-based timing attack. Experimental results show that the resulting protected implementations achieve identical cycle counts for all tested inputs, including (88) cycles in the three-function setting and (108) cycles in the five-function setting. At the same time, the numerical-error analysis indicates that the approximated nonlinear functions retain high accuracy. These results suggest that the proposed methodology provides a practical basis for constructing side-channel-resistant activation functions in embedded inference. --- ### Do Not Trust The Auctioneer: Learning to Bid in Feedback-Manipulated Auctions Authors: Luigi Foscari, Matilde Tullii, Vianney Perchet Published: 2026-05-21 Categories: stat.ML, cs.GT, cs.LG arXiv: https://arxiv.org/abs/2605.22438v1 Shilling is the use of artificial bids to make competition appear stronger and push prices upward. We study repeated first-price auctions in which shilling affects feedback but not allocation: the learner wins or loses against the real competing bid, but after a loss observes the maximum of the real bid and an independent shill bid. Thus the manipulation changes what the learner observes and hence how it learns to bid, without changing the outcome of the current auction. We analyze regret with respect to the best bid benchmark, assuming that the shill-bid distribution is known. Even then, shilling can mask the real bid, while useful side information appears only through intermittent low-shill events. Our algorithm combines a robust interval-elimination branch, which ignores the shilled report and achieves the dynamic-pricing rate $\tilde{\mathcal{O}}(T^{2/3})$, with an optimistic branch that debiases losing-side reports and exploits the resulting suffix information when it is reliable and achieves the first-price auctions rate $\tilde{\mathcal{O}}(\sqrt{T})$. A validation and racing procedure lets the algorithm use these optimistic updates without knowing the right scale or feedback geometry in advance. We complement the upper bounds with a matching lower bound, up to logarithmic factors, in the single-active-region case. Overall, the results show that even feedback-only shilling can sharply alter the statistical difficulty of repeated bidding. --- ### Characterizing the Fault Response of the Intel Neural Compute Stick 2 Under Single-Pulse Electromagnetic Fault Injection Authors: Štefan Kučerák, Jakub Breier, Xiaolu Hou Published: 2026-05-21 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.22437v1 Vision processing units and other commercial neural-network inference accelerators are increasingly deployed in safety-relevant edge applications, but their fault response under transient hardware disturbances remains poorly characterized in the open literature. For the Intel Movidius Myriad X, packaged as the Intel Neural Compute Stick 2 (NCS2), only a single feasibility study has been published. We report a systematic single-pulse electromagnetic fault injection (EMFI) campaign on the NCS2 running three ImageNet-trained convolutional neural networks (ResNet-18, ResNet-50, VGG-11) on the OpenVINO runtime. Across 1,536 spot-test trials at characterized hotspots and approximately 16,000 parameter-search trials, single pulses produce four reproducible outcome classes: no measured accuracy change, minor silent data corruption, major persistent degradation that survives across subsequent inferences until model reload, and device hangs requiring USB power-cycling; these outcomes are respectively interpreted as no-effect, SDC with possible SET-like or small persistent-state mechanisms, SEU-like persistent corruption, and SEFI-like loss of functionality. Two findings are central. First, the major-degradation class can be induced at 18-31% of trials at characterized hotspots, with post-collapse top-1 accuracy below five percent and persistence across all subsequent inferences until explicit model reload - a regime that no inference-API-level mechanism detects. Second, this regime is also inducible by pulses delivered to an idle device with the model already loaded, demonstrating that load-time integrity checks alone are insufficient. We discuss mitigation strategies graded by class, focusing on mechanisms implementable at the application level without modification to the device firmware or the OpenVINO runtime. --- ### Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation Authors: Genoveffa Martone, Helena Bonaldi, Marco Guerini Published: 2026-05-21 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.22435v1 Hate speech and misinformation frequently co-occur online, amplifying prejudice and polarization. Given their scale, using Large Language Models (LLMs) to assist expert counterspeech (CS) writing has gained interest, yet prior work has addressed these phenomena separately. We bridge this gap by studying CS generation in contexts where both hate and misinformation co-occur. We test three knowledge-driven generation strategies: first we prompt an LLM with fact-checkers' guidelines and fact-checking articles; secondly, with NGOs' guidelines and reports; thirdly, we create a mixed strategy that combines guidelines and documents from both. 23 experts revise the generated CS, which are assessed via human and automatic metrics. While LLMs produce adequate CS in 40% of cases, expert edits substantially improve naturalness, exhaustiveness, and adherence to guidelines. Based on the post-edited CS, the mixed strategy proves to be the most effective in crowdsourcing evaluation, pairing strong factual correction with stereotype mitigation and empathetic engagement. We release a dataset of hateful and misinformed claims with expert-verified CS and supporting knowledge. --- ### AMUSE: Anytime Muon with Stable Gradient Evaluation Authors: Jueun Kim, Baekrok Shin, Jihun Yun, Beomhan Baek, Minhak Song, Chulhee Yun Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22432v2 Modern deep learning commonly relies on AdamW with prescribed learning rate schedules, but recent works challenge both components: Schedule-Free optimization removes explicit schedules via iterate averaging, and Muon improves the update geometry by orthogonalizing momentum for matrix parameters. Despite Muon's strong empirical performance, its underlying mechanism remains partially understood. We study Muon through the river-valley loss landscape, where useful training progress occurs along a flat, low-curvature bulk subspace (the river), while high-curvature dominant directions form steep valley walls that induce oscillations. We empirically show that while Muon's orthogonalization accelerates river progress by increasing the bulk component, it also amplifies dominant-direction noise, causing oscillatory trajectories. Building on this, we propose Anytime MUon with Stable gradient Evaluation (AMUSE), which integrates Muon's rapid bulk progress with the stabilizing effect of Schedule-Free averaging. AMUSE uses a time-varying interpolation coefficient that initially evaluates gradients near the fast Muon sequence for rapid adaptation, then gradually shifts toward the stable averaged sequence to suppress valley-wall oscillations. As a result, AMUSE requires no learning rate schedules and supports anytime training. Across vision tasks and large language model pretraining, AMUSE consistently improves the performance-iteration Pareto frontier over (Schedule-Free) AdamW and Muon. --- ### Time-varying rPPG signal separation via block-sparse signal model Authors: Kosuke Kurihara, Yoshihiro Maeda, Daisuke Sugimura, Takayuki Hamamoto Published: 2026-05-21 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2605.22425v2 Remote photoplethysmography (rPPG) enables non-contact measurement of cardiac pulse signals by analyzing subtle color changes in facial videos. Nevertheless, extracting rPPG signals remains challenging because of their extremely weak signal strength and susceptibility to illumination noise. In this paper, we propose an rPPG signal extraction method that exploits the quasi-periodic characteristics of rPPG signals. Our approach models quasi-periodicity of the rPPG signal, which arises from the stable cardiac cycle, as a block-sparse structure in the time-frequency domain. To incorporate a block-sparse model and enable adaptive signal separation under illumination fluctuations, we construct a time-varying signal separation framework. Experiments using a public dataset demonstrate the effectiveness of our method. --- ### Moment-Reenacting: Inverse Motion Degradation with Cross-shutter Guidance Authors: Xiang Ji, Guixu Lin, Zhengwei Yin, Jiancheng Zhao, Yinqiang Zheng Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22423v2 Motion degradation, manifested as blur in global shutter (GS) images or rolling shutter (RS) distortion in RS counterparts, remains a fundamental challenge in computational imaging, especially under fast motion or low-light conditions. While prior works have treated blur decomposition and RS temporal super-resolution as separate tasks, this separation fails to exploit their intrinsic complementarity. In this paper, we propose a unified framework to invert motion degradation and reenact imaging moment by jointly leveraging the complementary characteristics of GS blur and RS distortion. To this end, we introduce a novel dual-shutter setup that captures synchronized blur-RS image pairs and demonstrate that this combination effectively resolves temporal and spatial ambiguities inherent in both modalities. For allowing flexible performance-cost trade-offs, we further extend this dual-shutter setup to a stereo Blur-RS configuration with a narrow baseline. In addition, we construct a triaxial imaging system to collect a real-world dataset with aligned GS-RS pairs and ground-truth high-speed frames, enabling robust training and evaluation beyond synthetic data. Our proposed network explicitly disentangles motion into context-aware and temporally-sensitive representations via a dual-stream motion interpretation module, followed by a self-prompted frame reconstruction stage. Extensive experiments validate the superiority and generalizability of our approach, establishing a new paradigm for realistic high-speed video reconstruction under complex motion degradations. Codes and more resources are available at https://jixiang2016.github.io/dualBR_site/. --- ### FastTab: A Fast Table Recognizer with a Tiny Recursive Module and 1D Transformers Authors: Laziz Hamdi, Amine Tamasna, Pascal Boisson, Thierry Paquet Published: 2026-05-21 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.22422v1 Table structure recognition (TSR) requires both table-level coherence (row/column counts, headers, spanning cells) and precise separator localization. We introduce FastTab, a grid-centric TSR model that avoids autoregressive HTML decoding by combining (i) a lightweight Tiny Recursive Module (TRM) for global reasoning and (ii) axial 1D Transformer encoders that capture long-range dependencies along rows and columns. The model predicts row/column counts, header rows, and separators to construct a grid, then infers rowspan/colspan using ROI-aligned cell features. Across four benchmarks (PubTabNet, FinTabNet, PubTables-1M, and SciTSR), FastTab achieves competitive structure recovery performance while operating at low-latency inference. We further study robustness under pixel-level anonymisation and show an extension to curved separators for camera-captured documents. The source code will be made publicly available at https://github.com/hamdilaziz/FastTab . --- ### Diffusion-guided Generalizable Enhancer for Urban Scene Reconstruction Authors: Henry Che, Jingkang Wang, Yun Chen, Ze Yang, Sivabalan Manivasagam, Raquel Urtasun Published: 2026-05-21 Categories: cs.CV, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2605.22420v1 Urban scene reconstruction from real-world observations has emerged as a powerful tool for self-driving development and testing. While current neural rendering approaches achieve high-fidelity rendering along the recorded trajectories, their quality degrades significantly under large viewpoint shifts, limiting the applicability for closed-loop simulation. Recent works have shown promising results in using diffusion models to enhance quality at these challenging viewpoints and distill improvements back into 3D representations. However, they often require costly per-scene optimization, and the distilled representations remain fragile and fail to generalize beyond limited synthesized views. To address these limitations, we propose GenRe, a novel diffusion-guided generalizable enhancer for urban scene reconstruction. GenRe takes as input any pretrained 3D Gaussian representation and fixes the deficiencies within a few minutes. By learning to distill generative priors across diverse scenes, GenRe produces robust and high-fidelity representation efficiently that generalizes reliably to challenging unseen viewpoints (e.g., lane change). Experiments show that GenRe outperforms existing methods in both quality and efficiency and benefits various downstream tasks, enabling robust and scalable sensor simulation for autonomous driving. --- ### The Neglected Baseline in Model Interpretation Authors: Yongjin Cui, Xiaohui Fan Published: 2026-05-21 Categories: cs.CV, cs.SE arXiv: https://arxiv.org/abs/2605.22417v6 We observe that existing model interpretation methods generally ignore the baseline, and such neglect often results in imprecise or even incorrect interpretation. In this paper, we reformulate the task of model interpretation and the interpretation principles for model interpretation results to demonstrate the importance of the baseline. We further unify gradient-based methods, Integrated Gradients (IG) methods, and Taylor expansion, clarifying the connections among them and explicitly identifying the baseline for each method. On this basis, we analyze the flaws and errors in related model interpretation methods (IG, LayerCAM, ODAM, Difference Map). We advocate evaluating the quality of model interpretation results precisely through the attribution error between the attribution result and the attribution target, rather than adopting flawed evaluation methods, such as those based on marginal-effect or the assumption of perfect model performance. We revise IG and develope a model interpretation method with a clear and reasonable baseline, achieving better results. Our method supports model interpretation based on features from any layer. Interpretation based on features from different layers are all reasonable, and the differences among these results reflect varying degrees of feature extraction at different feature extraction stages. --- ### The Neglected Baseline in Model Interpretation Authors: Yongjin Cui, Xiaohui Fan Published: 2026-05-21 Categories: cs.CV, cs.SE arXiv: https://arxiv.org/abs/2605.22417v4 We observe that existing model interpretation methods generally ignore the baseline, and such neglect often results in imprecise or even incorrect interpretation. In this paper, we reformulate the task of model interpretation and the interpretation principles for model interpretation results to demonstrate the importance of the baseline. For the first time, we unify gradient-based methods, Integrated Gradients (IG), and Taylor expansion, clarify the relationships among the three, and explicitly identify the corresponding baseline for each method. This may have a significant impact on the further performance improvement of some gradient-based schemes. On this basis, we analyze the flaws and errors in related model interpretation methods (IG, LayerCAM, ODAM, Difference Map). We advocate evaluating the quality of model interpretation results precisely through the attribution error between the attribution result and the attribution target, rather than adopting flawed evaluation methods, such as those based on marginal-effect or the assumption of perfect model performance. We revise IG and develope a model interpretation method with a clear and reasonable baseline, achieving better results. Our method supports model interpretation based on features from any layer. Interpretation based on features from different layers are all reasonable, and the differences among these results reflect varying degrees of feature extraction at different feature extraction stages. --- ### Asymmetric Virtual Memory Paging for Hybrid Mamba-Transformer Inference Authors: An Xuan Nguyen Published: 2026-05-21 Categories: cs.LG, cs.DC, cs.PF arXiv: https://arxiv.org/abs/2605.22416v1 Hybrid language models like Jamba mix attention layers with State Space Models (SSMs), creating two memory cache types with opposite profiles: Key-Value (KV) caches grow linearly with sequence length, while SSM states stay fixed per layer. Current inference engines handle this poorly. Unified pools pad SSM states to attention page sizes, wasting up to 7.3x capacity. Static dual pools cannot adapt when prompt distributions shift between requests. We present Asymmetric Virtual Memory Paging (AVMP). The allocator separates the two cache types into physically distinct pools behind a unified virtual address space, and migrates capacity between pools when one runs out. Migration triggers only on allocation failure, keeping behavior deterministic. We evaluate AVMP across 270 synthetic cells plus 60 cells of ShareGPT trace replay on an RTX 3060 12GB. Out-of-Memory events drop 7.6% and request throughput improves 1.83x to 13.3x across synthetic workloads and 2.36x on ShareGPT. All gains hold under paired-bootstrap 95% confidence intervals. A phase-time breakdown reveals two distinct mechanisms: shorter OOM recovery on capacity-pressured workloads, and faster allocation calls on KV-heavy workloads. Implementation is pure Python; Triton integration is future work. --- ### Towards Clinically Interpretable Ophthalmic VQA via Spatially-Grounded Lesion Evidence Authors: Xingyue Wang, Bo Liu, Meng Wang, Zhixuan Zhang, Chengcheng Zhu, Huazhu Fu, Jiang Liu Published: 2026-05-21 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.22414v1 Visual Question Answering (VQA) holds great promise for clinical support, particularly in ophthalmology, where retinal fundus photography is essential for diagnosis. However, ophthalmic VQA benchmarks primarily emphasize answer accuracy, neglecting the explicit visual evidence necessary for clinical interpretability. In this work, we introduce FundusGround, a new benchmark for clinically interpretable ophthalmic VQA with spatially-grounded lesion evidence. Specifically, we propose a three-stage pipeline that collects 10,719 fundus images with 15,595 image-level meticulously annotated lesions. To ensure anatomical consistency and clinical validity, all lesions are spatially localized using the Early Treatment Diabetic Retinopathy Study (ETDRS) grid, enabling standardized mapping to nine clinically meaningful retinal regions. Built upon this structured lesion evidence, 72,706 questions are then generated spanning four formats: open-ended, closed-ended, single-choice, and multiple-choice. We further benchmark multiple general- and medical- large vision-language models using dual metrics for answer accuracy and lesion-level reasoning. The experiments demonstrate that incorporating lesion-level visual evidence consistently improves model performance and transparency, highlighting the necessity of explicit spatial grounding for reliable and explainable ophthalmic VQA. --- ### From Recognition to Reasoning: Benchmarking and Enhancing MLLMs on Real-World Receipt Document Understanding Authors: Yandi Wang, Libin Zhan, Ziwei Huang, Tiancheng Luo, Yuxuan Jiang, Wang Dong, Leilei Gan, Jun Chen Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22413v1 Extracting structured information from visual documents (Visual Information Extraction, VIE) is a cornerstone of business automation. While recent Multimodal Large Language Models (MLLMs) have shown promising capabilities, existing benchmarks suffer from critical limitations in scale and realism, lack semantic granularity, and fail to cover diverse document types. To bridge this gap, we introduce ReceiptBench, a large-scale, human-annotated benchmark consisting of 10k diverse receipts, organizing information extraction into four hierarchical sub-tasks: (1) Basic Perception for raw text spotting, (2) Format Normalization for strictly following standardization instructions, (3) Semantic Reasoning for inferring implicit attributes from context, and (4) Structure Parsing for handling nested line items. Furthermore, we propose a two-stage training framework incorporating Metric-Aware Group Relative Policy Optimization (GRPO), which translates rigorous evaluation constraints into reinforcement learning signals to enhance structural consistency. Extensive experiments demonstrate that our method yields state-of-the-art performance, surpassing leading proprietary models on complex reasoning tasks. We release our datasets and code at https://github.com/wwwT0ri/ReceiptBench. --- ### DeferMem: Query-Time Evidence Distillation via Reinforcement Learning for Long-Term Agent Memory Authors: Jianing Yin, Tan Tang, Yingcai Wu Published: 2026-05-21 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.22411v2 Large language model (LLM) agents still struggle to effectively use long-term memory, with answer-supporting evidence often scattered across long conversational histories and buried in substantial irrelevant content. Existing memory systems commonly process memory into dedicated units before future queries are known and retrieve these preconstructed units at query time. Because the contents and granularity of these units are determined before the current query is known, the retrieved units can remain coarse and noisy for that query, leaving downstream answerers to further denoise them and uncover query-specific evidence. We present DeferMem, a long-term memory framework that decouples this problem into high-recall candidate retrieval and query-conditioned evidence distillation. DeferMem uses a lightweight segment-link structure to organize raw history and retrieve broad candidates at query time. A memory distiller then distills these high-recall but highly noisy candidates into a set of faithful, self-contained, and query-conditioned evidence. To train this distiller, we introduce DistillPO, a reinforcement learning algorithm that formulates post-retrieval evidence distillation as a structured action comprising message selection and evidence rewriting. It optimizes this action with a decomposed-and-gated reward pipeline and structure-aligned advantage assignment, gating reward components from validity to quality checks while exposing answerability-related feedback early and assigning each reward to its responsible output span. On LoCoMo and LongMemEval-S, DeferMem surpasses strong baselines in QA accuracy and memory-system efficiency, achieving the highest QA accuracy and fastest runtime while consuming no commercial-API tokens for memory operations. --- ### DeferMem: Query-Time Evidence Distillation via Reinforcement Learning for Long-Term Memory QA Authors: Jianing Yin, Tan Tang Published: 2026-05-21 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.22411v1 Large language model (LLM) agents still struggle with long-term memory question answering, where answer-supporting evidence is often scattered across long conversational histories and buried in substantial irrelevant content. Existing memory systems typically process memory before future queries are known, then retrieve the resulting units based on similarity rather than their utility for answering the query. This workflow leaves downstream answerers to denoise retrieved candidates and reconstruct query-specific evidence. We present DeferMem, a long-term memory framework that decouples this problem into high-recall candidate retrieval and query-conditioned evidence distillation. DeferMem uses a lightweight segment-link structure to organize raw history and retrieve broad candidates at query time. It then applies a memory distiller trained with DistillPO, our reinforcement learning algorithm for distilling the high-recall but highly noisy candidates into a set of faithful, self-contained, and query-conditioned evidence. DistillPO formulates post-retrieval evidence distillation as a structured action comprising message selection and evidence rewriting. It optimizes this action with a decomposed-and-gated reward pipeline and structure-aligned advantage assignment, gating reward components from validity to quality checks while exposing task-level correctness feedback early and assigning each reward to its responsible output span. On LoCoMo and LongMemEval-S, DeferMem surpasses strong baselines in QA accuracy and memory-system efficiency, achieving the highest QA accuracy with the fastest runtime and zero commercial-API token cost for memory operations. --- ### Minimum Description Length based Granular-Ball Tree Regularization for Spectral Clustering Authors: Zeqiang Xian, Caihui Liu, Yong Zhang, Wenjing Qiu Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22410v3 This paper proposes Minimum Description Length based Granular-Ball Tree Regularized Spectral Clustering (MDL-GBTRSC), a fixed-K method that uses granular balls as regional descriptors for sample-level affinity construction rather than as reduced graph nodes. A reciprocal-neighborhood continuity term is incorporated into an MDL-based refinement criterion to construct a granular-ball hierarchy, and the structural scales of the terminal regions are transferred to the original sample graph before spectral partitioning. Experiments on 18 real and 20 synthetic datasets, comprising 109,969 samples in total, compare MDL-GBTRSC with nine competing methods. On the 17 real datasets completed by all methods, MDL-GBTRSC achieves mean ARI and NMI values of 0.5998 and 0.6613, compared with the next-highest means of 0.5792 and 0.6285, respectively. On the 20 synthetic datasets, it achieves mean ARI and NMI values of 0.9618 and 0.9688, compared with the next-highest means of 0.7921 and 0.8306, and is best or tied on 15 datasets for both metrics. These results indicate that MDL-selected regional structure provides useful local information for affinity construction while preserving the resolution of the original spectral graph. Source code and experimental materials are available at https://github.com/AnonymousUser0816/MDL-GBTRSC. --- ### Minimum Description Length based Granular-Ball Tree Regularization for Spectral Clustering Authors: Zeqiang Xian, Caihui Liu, Yong Zhang, Wenjing Qiu Published: 2026-05-21 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.22410v2 Spectral clustering largely depends on the affinity graph, yet constructing a graph that preserves reliable local connectivity while adapting to heterogeneous data structures remains challenging. Existing granular-ball-based spectral clustering methods usually reduce graph complexity by using coarse-grained representatives. However, the learned local regions are often treated as graph nodes or anchors, and their structural information is not sufficiently used to regularize the original sample-level graph. To address this issue, this paper proposes a Minimum Description Length based Granular-Ball Tree-Regularized Spectral Clustering method, termed MDL-GBTRSC. The proposed method constructs a granular-ball tree through local MDL model selection, with reciprocal neighborhood continuity used to discourage splits that break reliable local connections. The stable leaf balls obtained from the tree provide coding-scale information for regularizing the sample-level affinity graph. In addition, a shared-neighbor bridge code is introduced to adjust weak local bridge relations without requiring an additional user-specified threshold. In this way, MDL-GBTRSC connects interpretable local representation learning with affinity graph construction in a unified spectral clustering framework. Experiments on real and synthetic datasets show that MDL-GBTRSC achieves the best average ARI and NMI under the adopted fixed-configuration protocol compared with classical spectral clustering baselines and representative granular-ball, micro-cluster, and anchor-based methods. --- ### Translating Signals to Languages for sEMG-Based Activity Recognition Authors: Ming Wang, Haoxuan Qu, Qiuhong Ke, Wei Zhou, Hossein Rahmani, Jun Liu Published: 2026-05-21 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.22403v1 Surface electromyography (sEMG) signal-based activity recognition has attracted increasing research attention in recent years. To develop accurate sEMG signal-based activity recognizers, numerous approaches have been proposed. Some studies focus on designing larger and more expressive model architectures to enhance the representational capacity of sEMG signals, while others aim to enrich model priors through large-scale pretraining, thereby improving recognition performance. Recently, large language models (LLMs) have shown remarkable generalization and reasoning capabilities in natural language processing, whose implicit knowledge, learned from extensive linguistic descriptions of actions, opens new possibilities for interpreting sEMG signals and inferring activity intentions. Motivated by this, we propose LLM-sEMG, a novel framework that leverages LLMs as sEMG activity recognizers. Within this framework, we design a language-oriented mapping mechanism that converts continuous sEMG sequences into sEMG language, integrating several strategies to further facilitate the signal-to-language mapping process. Extensive experiments demonstrate that the proposed framework achieves highly accurate sEMG signal-based activity recognition using large language models. --- ### Cross-Species RSA Reveals Conserved Early Visual Alignment but Divergent Higher-Area Rankings Across Human fMRI and Macaque Electrophysiology Authors: Nils Leutenegger Published: 2026-05-21 Categories: cs.LG, cs.NE, q-bio.NC arXiv: https://arxiv.org/abs/2605.22401v2 CORRECTION (August 2026): an evaluation-mode defect in the shared feature-extraction pipeline affected the predictive-coding and STDP conditions. It applies to both sides of every comparison here: the human values are reprinted from the companion study and the macaque values use the same checkpoints. In a single-seed re-evaluation with repaired checkpoints, macaque STDP at V1 moves from 0.305 to 0.266 and PC from 0.210 to 0.178, level with the untrained baseline, so finding (2) holds for STDP but not for PC; the V4 and IT orderings also change, while V2 is unchanged. Findings (1), (3) and (4) are unaffected, and the other conditions change by at most 0.0034. The five-seed analysis, Kendall's tau, noise ceilings and stimulus control have not been re-computed. See the correction note on page 1; the original abstract below is unchanged from v1. Follow-up study: arXiv:2608.12408. Does the relationship between learning rules and brain alignment generalize across species? We test the same five learning rules against macaque electrophysiology. The macaque data come from MajajHong2015 (V4/IT, 3,200 presentations, 88/168 neurons) and FreemanZiemba2013 (V1/V2, 135 stimuli, 102/103 neurons). Using RSA with identical model weights from our human study, we find: (1) all models achieve higher alignment with macaque early visual cortex (rho = 0.15-0.30 at V1/V2) than with human fMRI (rho = 0.01-0.08); (2) STDP and PC produce the highest macaque V1/V2 alignment; (3) at IT, rankings show no detectable correlation across species (Kendall's tau = 0.00), though this null is expected given that n = 5 provides power only at tau = +/-1.0; (4) a pretrained ResNet-50 achieves rho = 0.25 at macaque IT, substantially above all custom CNN conditions (rho = 0.07-0.14), suggesting IT alignment is limited by model capacity rather than by the learning rule. ---